Overview

This policy outlines the principles for the organization, storage, and management of scientific data generated under Horizon Europe Grant Project NAR-SAR-IPH-101087403.

The established principles are rooted in FAIR data principles and focus on:

• Enhancing data findability and accessibility for both human users and automated systems.

• Ensuring consistency across all datasets to facilitate reproducibility and reusability.

• Providing clear pathways for structured storage and retrieval.

• Enabling metadata to be easily previewed and interpreted.

General Principles

• Data and metadata must be stored in interoperable, non-proprietary formats (e.g. TIF for hyperspectral datacubes, CSV for tabular metadata).

• Original versions outputted by the instruments may also be included.

• The file hierarchy must remain consistent across same-type experiments to facilitate data retrieval by both humans and machines.

• Metadata should be stored in easily accessible and structured formats (e.g., CSV) to enhance usability and previewability without requiring archive (e.g., ZIP) extraction.

• Paths and folder structures should be clearly defined and maintained to enable reproducibility.

Data and Metadata Guidelines

Data and metadata follow two levels of schema:

• Repository-level schema, facilitated by Zenodo

• Dataset-level schema, defined by internal guidelines

Dataset-level guidelines apply to same-type, domain-specific data and metadata. Each type of domain-specific guidelines are formulated separately, uploaded to Zenodo as an individual entry, and linked to the records which follow those guidelines.

A dataset may contain different types of domain specific data and metadata.

All versions of dataset-level guidelines are available below. Datasets should be organized according to the latest available version of dataset-level schema.

Hyperspectral Imagery